In humans, hereditary cerebellar ataxias develop gradually resulting from the degeneration of cerebellar neurons and their afferent and efferent connections. The long term goal of this investigation is to see whether insulin-like growth factor I (IGF-1) can rescue cerebellar neurons from dying in cerebeltar mutant mice, thereby evaluating the therapeutic potential of IGF-l in treating cerebellar ataxia in humans. IGF-l is an anabolic growth factor required for optimal neuronal proliferation, differentiation and survival. IGF-l's neurotropic effect is best illustrated during the development of the cerebellum, where IGF-l and its receptor genes are normally expressed coordinately with the postnatal cerebellar growth spurt. When cerebellar growth is affected by gene mutations (wv and pcd mice), IGF-l's biological activity usually decreases before ataxia occurs, suggesting that normal IGF-l levels are pivotal for the functional integrity of cerebellar cytoarchitecture. On the other hand, IGF-l transgenic mice have bigger brains with more myelin and more brain cells. Among all brain regions, the cerebellum is affected most. It is twice the normal size, containing 92 percent more granule cells and 20 percent more Purkinje cells than are found in wild type littermates. To fully evaluate the therapeutic potential of IGF-l in the treatment of cerebellar ataxia, this investigation will: 1. characterize the cellular mechanism of IGF-l's neuroprotection for cerebellar neurons; 2. cross breed IGF-l transgenic mice with wv and pcd mutant mice and examine the resulting histology, molecular biology and behavior changes in IGF-I transgenic mice that contain zero, one or two wv or pcd alleles; and 3. evaluate the therapeutic effects of IGF-l delivered by microencapsulated mammalian cells engineered to synthesize and release IGF-l upon stimulation. The results of this investigation will provide crucial information about the therapeutic potential of IGF-I in treating hereditary cerebellar ataxia.